JP5397387B2 - Fuel cell system - Google Patents

Description

  The present invention relates to a fuel cell system for estimating the water content of a polymer electrolyte fuel cell.

As is widely known in the past, in order to efficiently generate power in a polymer electrolyte fuel cell, it is desirable that the electrolyte membrane be in a moderately wet state so that the amount of water in the fuel cell is not excessive or insufficient. . As a technique for controlling the amount of water in the cell surface of a fuel cell, for example, one described in Patent Document 1 is known. This Patent Document 1 adjusts at least one of pressure loss characteristics depending on the pressure, humidity, temperature, flow rate, and flow path shape of a reaction gas (a general term for an oxidizing gas typified by air and a fuel gas typified by hydrogen gas). Thus, it is disclosed to control the distribution of the amount of water as droplets or water vapor in the cell surface.
JP 2004-335444 A

  However, in an actual single cell, there is moisture movement through the electrolyte membrane between the anode electrode and the cathode electrode. In this regard, Patent Document 1 does not consider water movement between the electrodes, and it is difficult to accurately estimate and control the distribution of moisture in the cell plane.

  Further, Patent Document 1 does not disclose any estimation and control of the water content distribution while the fuel cell system is left after being stopped. In fact, during the standing of the fuel cell system, water accumulated on the cathode electrode side due to the water vapor partial pressure difference (temperature difference) moves to the anode electrode side through the electrolyte membrane. As a result, when the fuel cell system is next started, there are cases where the amount of residual water in the anode electrode side channel is large and the pressure loss is high. For this reason, when restarted, supply of hydrogen gas to the anode electrode is insufficient, and the MEA of the single cell may be deteriorated.

  An object of the present invention is to provide a fuel cell system capable of improving the estimation accuracy of the amount of moisture being left.

In order to achieve the above object, a fuel cell system of the present invention comprises an anode electrode, a cathode electrode, an electrolyte membrane between them, a fuel gas flow path for supplying fuel gas to the anode electrode, and an oxidizing gas supplied to the cathode electrode A fuel cell in which a plurality of single cells each having an oxidizing gas flow path are stacked, water movement performed between the anode electrode and the cathode electrode via the electrolyte membrane, and from the stop to restart of the fuel cell system In consideration of the temperature drop profile of each single cell obtained by calculating the temperature of each single cell multiple times during the standing period, the fuel gas flow path in the cell surface of each single cell during the standing period an estimation unit for estimating a residual water content distribution, when the total amount of the residual water content in the fuel gas flow path which is grasped from said estimated residual water content distribution reaches a predetermined threshold value, to the fuel gas flow passage And operation control unit for executing air treatment, those having a.

According to the present invention, it is possible to improve since the consideration of the water transfer between electrodes, in the cell lamination direction not in the cell surface only the estimation accuracy of residual water distribution. Further, since also consider lowering profile of each single cell obtained during the unused period, Ru can be accurately grasped water movement during the unused period. That is, the present inventors have found that the temperature distribution between single cells in the cell stacking direction differs depending on the temperature profile, and the amount of water movement to the fuel gas flow path due to the temperature difference between the electrodes differs. However, it is possible to estimate the transition of the temperature distribution between the single cells by acquiring and considering the temperature drop profile during the leaving period as in the present invention, and the fuel gas flow path during the leaving period can be estimated . The amount of remaining water can be estimated accurately. As a result of the estimation, when the remaining water amount (total amount) reaches a predetermined threshold value, the scavenging process is executed as a countermeasure control , so that the remaining water in the fuel gas channel can be reduced, resulting in the system It is possible to suppress the occurrence of water clogging in the fuel gas flow path during leaving. Therefore, fuel gas supply shortage can be suppressed at the time of system restart. In addition, when the scavenging process is performed when the system is restarted or the system is stopped, the time can be shortened.
Note that the estimation unit preferably estimates at least one of the residual water amount distribution of the oxidizing gas flow and the water content distribution of the electrolyte membrane in the cell plane of each single cell in consideration of the water movement.

  In the following description, the stop, leave, and restart of the fuel cell system may be abbreviated as “system stop”, “system leave”, and “system restart”, respectively.

  Preferably, the fuel cell system further includes a temperature sensor that detects a temperature related to the fuel cell, and the estimation unit may calculate the temperature of each single cell using one temperature detected by the temperature sensor.

  Thereby, the temperature information of each single cell can be acquired while reducing the number of parts and the cost as compared with the case where temperature sensors are individually provided for all the single cells.

  Preferred embodiments of the present invention will be described below with reference to the accompanying drawings. First, an outline of a fuel cell system including a fuel cell and a moisture amount estimation device thereof will be described, and then an estimation of the moisture amount of the fuel cell and a control example using the estimation will be described. Hereinafter, hydrogen gas will be described as an example of fuel gas, air will be described as an example of oxidizing gas, and fuel gas and oxidizing gas may be collectively referred to as reaction gas.

A. Overview of Fuel Cell As shown in FIGS. 1 and 2, a fuel cell 1 having a stack structure has a cell laminate 3 formed by laminating a plurality of solid polymer electrolyte type single cells 2. Current collector plates 5a and 5b, insulating plates 6a and 6b, and end plates 7a and 7b are arranged outside the single cells 2 (hereinafter referred to as “end cell 2a”) at both ends of the cell stack 3, respectively. The Tension plates 8, 8 are bridged between end plates 7a, 7b and fixed with bolts 9, and an elastic module 10 is provided between the end plate 7b and the insulating plate 6b.

  Hydrogen gas, air and refrigerant are supplied to the manifold 15a in the cell stack 3 from the supply pipe 14 connected to the supply ports 11a, 12a and 13a of the end plate 7a. Thereafter, hydrogen gas, air, and refrigerant flow in the plane direction of the single cell 2, reach the manifold 15b in the cell stack 3, and fuel from the discharge pipe 16 connected to the discharge ports 11b, 12b, and 13b of the end plate 7a. It is discharged out of the battery 1. The supply pipe 14, the manifolds 15a and 15b, and the discharge pipe 16 are provided corresponding to the fluids of hydrogen gas, air, and refrigerant, but in FIG. .

  As shown in FIG. 3, the single cell 2 includes an MEA 20 and a pair of separators 22A and 22B. The MEA 20 (membrane-electrode assembly) includes an electrolyte membrane 23 made of an ion exchange membrane, and an anode electrode 24A and a cathode electrode 24B sandwiching the electrolyte membrane 23. The electrode 24A faces the hydrogen flow path 25A of the separator 22A, and the electrode 24B faces the air flow path 25B of the separator 22B. In addition, the refrigerant channels 26A and 26B of the separators 22A and 22B communicate between the adjacent single cells 2 and 2.

  FIG. 4 is a plan view of the separator 22A. The separator 22A has a hydrogen inlet 27a, an air inlet 28a, a refrigerant inlet 29a, a hydrogen outlet 27b, an air outlet 28b, and a refrigerant outlet 29b that are formed through the hydrogen channel 25A. The inlets 27a, 28a, and 29a constitute a part of the manifold 15a corresponding to each fluid, and similarly, the outlets 27b, 28b, and 29b constitute a part of the manifold 15b corresponding to each fluid.

  In the separator 22A, hydrogen gas is introduced from the inlet 27a into the hydrogen flow path 40 and discharged to the outlet 27b. This also applies to the refrigerant flow. Although not described in detail, air also flows in the planar direction of the separator 22B configured similarly to the separator 22A. In this way, hydrogen gas and air are supplied to the electrodes 24A and 24B in the single cell 2, thereby causing an electrochemical reaction in the MEA 20 and obtaining an electromotive force. In addition, this electrochemical reaction generates water and generates heat on the electrode 24B side. And the heat | fever of each single cell 2 is reduced because a refrigerant | coolant flows.

  5A to 5C are schematic plan views showing other channel shapes of the separator to which the present embodiment can be applied. Instead of the form of the straight groove flow path (repetitive irregularities extend in one direction) shown in FIG. 4, the flow path shapes of the flow paths 25A, 25B, 26A, and 26B are changed in the middle as shown in FIG. 5A. A serpentine channel shape having a folded portion at the end can also be used. Further, as shown in FIG. 5B, the flow paths 25A, 25B, 26A, and 26B can be wave-shaped, or as shown in FIG. Further, with regard to the flow of the reaction gas, instead of the co-flow type (hydrogen gas and air flow in the same direction) as understood from FIGS. 1 and 4, hydrogen gas and air flow in opposite directions. Counter flow type can also be adopted. Further, the orientation of the separators 22A and 22B may be either vertical or horizontal. That is, the estimation of the moisture content of the fuel cell 1 to be described later is not limited to the hardware configuration of the fuel cell 1.

B. Overview of Fuel Cell System As shown in FIG. 6, the fuel cell system 100 includes an air piping system 300, a hydrogen piping system 400, a refrigerant piping system 500, and a control device 600. The fuel cell system 100 can be mounted on various moving bodies such as vehicles, ships, airplanes, and robots, and can also be applied to stationary power sources. Here, a fuel cell system 100 mounted on an automobile will be described as an example.

  The air piping system 300 supplies and discharges air to and from the fuel cell 1 and includes a humidifier 30, a supply channel 31, a discharge channel 32, and a compressor 33. Air in the atmosphere (air in a low wet state) is taken in by the compressor 33 and is pumped to the humidifier 30, and moisture is exchanged with the oxidizing off gas in a high wet state in the humidifier 30. As a result, moderately humidified air is supplied from the supply channel 31 to the fuel cell 1. The discharge passage 32 is provided with a back pressure valve 34 that adjusts the air back pressure of the fuel cell 1. Further, in the vicinity of the back pressure valve 34, a pressure sensor P1 for detecting air back pressure is provided. The compressor 33 is provided with a flow rate sensor F1 that detects an air supply flow rate to the fuel cell 1.

  The hydrogen piping system 400 supplies and discharges hydrogen gas to and from the fuel cell 1, and includes a hydrogen supply source 40, a supply flow path 41, a circulation flow path 42, a shut valve 43, and the like. The hydrogen gas from the hydrogen supply source 40 is decompressed by the regulator 44, and then the flow rate and pressure are adjusted with high accuracy by the injector 45. Thereafter, the hydrogen gas merges with the hydrogen off-gas pumped by the hydrogen pump 46 on the circulation flow path 42 at the merge point A and is supplied to the fuel cell 1. A purge passage 47 with a purge valve 48 is branchedly connected to the circulation passage 42, and the hydrogen off-gas is discharged to the discharge passage 32 by opening the purge valve 48. A pressure sensor P <b> 2 that detects the supply pressure of hydrogen gas to the fuel cell 1 is provided downstream of the junction A. The hydrogen pump 46 is provided with a flow rate sensor F2. In another embodiment, the fuel off-gas may be introduced into a hydrogen diluter or the like, and a gas-liquid separator may be provided in the circulation channel 42.

  The refrigerant piping system 500 circulates and supplies a refrigerant (for example, cooling water) to the fuel cell 1, and includes a cooling pump 50, a refrigerant channel 51, a radiator 52, a bypass channel 53, and a switching valve 54. The cooling pump 50 pumps the refrigerant in the refrigerant flow path 51 into the fuel cell 1. The refrigerant flow path 51 includes a temperature sensor T1 on the refrigerant inlet side of the fuel cell 1 and a temperature sensor T2 on the refrigerant outlet side of the fuel cell 1. The radiator 52 cools the refrigerant discharged from the fuel cell 1. The switching valve 54 is constituted by, for example, a rotary valve, and switches the refrigerant flow between the radiator 52 and the bypass flow path 53 as necessary.

  The control device 600 is configured as a microcomputer having a CPU, a ROM, and a RAM inside. Detection information of sensors (P1, P2, F1, F2, T1, T2) for detecting the pressure, temperature, flow rate and the like of the fluid flowing through the piping systems 300, 400, 500 is input to the control device 600. In addition to the detection information of the current sensor 61 that detects the current value generated by the fuel cell 1, detection information such as the outside air temperature sensor 62, the vehicle speed sensor 63, and the accelerator opening sensor is input to the control device 600. The control device 600 controls various devices (the compressor 33, the shut valve 43, the injector 45, the hydrogen pump 46, the purge valve 48, the cooling pump 50, the switching valve 54, etc.) in the system 100 according to the detection information and the like. The overall operation of the fuel cell system 100 is controlled. The control device 600 reads various detection information, and estimates the moisture content of the fuel cell 1 using various maps stored in the ROM.

  As illustrated in FIG. 7, the control device 600 estimates a moisture amount of the fuel cell 1 and implements control based on the moisture amount, and includes a storage unit 65, a detection unit 66, an estimation unit 67, and an operation control unit 68. Is provided. The storage unit 65 stores various programs and various maps for realizing estimation and control of the water content of the fuel cell 1. The map is obtained in advance by experiment or simulation. The detection unit 66 reads detection information such as various sensors (P1, P2, F1, F2, T1, T2, 61 to 63). The operation control unit 68 transmits a control command to various devices based on the estimation result by the estimation unit 67, and controls the operation so that the fuel cell 1 is in a desired operation state (for example, a moisture state, a temperature state, etc.). . At this time, the operation control unit 68 executes control in which the anode side and the cathode side are distinguished as necessary.

  The estimation unit 67 estimates the water content of the fuel cell 1 with reference to various maps in the storage unit 65 based on the information acquired by the detection unit 66. Specifically, the estimation unit 67 estimates the residual water amount distribution and the water content distribution in the cell plane of the single cell 2 in consideration of the water movement performed between the electrodes 24A and 24B via the electrolyte membrane 23. The estimation unit 67 also estimates a residual water amount distribution and a water content distribution in the stacking direction of each single cell 2 (hereinafter referred to as the cell stacking direction).

  Here, “in the cell plane” means the inside of the single cell 2 in the plane direction of the single cell 2 (a direction parallel to the paper surface of FIG. 4 and a direction perpendicular to the cell stacking direction). “Residual water amount” means the amount of liquid water present in the reaction gas flow path of the single cell 2. The reactive gas flow channel is a general concept of the hydrogen flow channel 25A and the air flow channel 25B. “Water content” means the amount of water contained in the electrolyte membrane 23 of the single cell 2.

C. Method for Estimating Moisture Content of Fuel Cell In the moisture content estimation method of the present embodiment, the residual water amount and the water content are estimated separately, and at that time, the residual water amount distribution is estimated separately for the anode side and the cathode side. Further, regarding the residual water content and the water content, not only the distribution in the cell plane but also the distribution in the cell stacking direction is estimated. Below, the estimation method of the water distribution (residual water amount distribution and water content distribution) in the cell plane will be described first. Next, how to consider the temperature variation / distribution variation in the cell stacking direction for estimation will be described, and a method for estimating the water distribution in the cell stacking direction will be described.

1. Method for Estimating Water Distribution in Cell Surface As shown in FIG. 8, first, current value I, cell inlet temperature T _{in, i} , cell outlet temperature T _{OUT, i} , air flow rate Q _{air, i} , hydrogen flow rate Q _{H2 , I} , air back pressure P _{air, i} and hydrogen pressure PH _{2, i} are read (step S1).

Here, the current value I is detected by the current sensor 61. The subscript “i” _{in the} cell inlet temperature T _{in, i} or the like is a cell channel indicating the position of the single cell 2 in the cell stack 3. Specifically, when the cell stack 3 shown in FIG. 9 is modeled, a reaction gas supply port (corresponding to the supply ports 11a and 12a in FIG. 1) and a discharge port (the discharge ports 11b and 12b in FIG. 1). The cell channel “i” of the end cell 2a closest to (1) is 1. When 200 unit cells 2 are stacked, the cell channel “i” of the other end cell 2 a is 200.

The cell inlet temperature T _{in, i} and the cell outlet temperature T _{OUT, i} are the refrigerant temperatures at the refrigerant inlet 29a and the refrigerant outlet 29b of the single cell 2 (cell channel: i), respectively. The air flow rate Q _{air, i} and the hydrogen flow rate Q _{H2, i} are supply flow rates of air and hydrogen gas flowing into the air inlet 28a and the hydrogen inlet 27a of the single cell 2 _i , respectively. The air back pressure P _{air, i} and the hydrogen pressure PH _{2, i} are the pressures of air and hydrogen gas at the air outlet 28b and the hydrogen inlet 27a of the single cell 2 _i , respectively. When the fuel cell has only one single cell 2 or when temperature variation / distribution variation in the cell stacking direction is not taken into consideration, the situation is as follows.
T _{in, i} : detected value by temperature sensor T 1 T _{OUT, i} : detected value by temperature sensor T 2 Q _{air, i} : detected value by flow sensor F 1 Q _{H 2, i} : hydrogen supply flow rate obtained from detected value by flow sensor F 2 P _{air, i} : detected value by pressure sensor P 1 P _{H2, i::} detected value by pressure sensor P ₂

On the other hand, when the fuel cell 1 has a plurality of single cells 2, the amount of heat release, pressure loss, and the like differ depending on the position in the cell stacking direction. There is. Therefore, it is desirable to use the cell inlet temperature Tin _{, i,} etc. _in consideration of this point. This consideration will be described later.

It should be noted that each detected value used as the cell inlet temperature T _{in, i} or the like can be a value other than the above sensor or a value obtained by a calculation method. In other words, the temperature sensor, the flow rate sensor, and the pressure sensor may be provided at positions other than those shown in FIG. 6, and the number and positions can be appropriately changed in design. For example, a hydrogen flow sensor may be provided near the hydrogen supply port 11a of the fuel cell 1, and the detected value may be used as the hydrogen flow _{QH2, i} . Further, the cell inlet temperature T _{in, i} and the cell outlet temperature T _{OUT, i} can be estimated by attaching a temperature sensor to the end cell 2a or the end plates 7a, 7b. Thus, by measuring the temperature of the fuel cell stack itself instead of the temperature of the refrigerant, it is possible to estimate the moisture with higher accuracy.

At step S2 shown in FIG. 8, the cell inlet temperature _{T in,} cathode inlet dew point of each single cell _{2 i} from _{i T d, CA} and the anode inlet dew point _{T d,} and calculates the _AN. In the present embodiment, since the humidifier 30 is used in the fuel cell system 1, the cell inlet temperature T _{in, i} can be used as the cathode inlet dew point T _{d, CA} and the anode inlet dew point T _{d, AN} , respectively. . That is, when the air inlet 28a and the hydrogen inlet 27a are close to the refrigerant inlet 29a, it can be expressed as follows, and the stacking variation of the dew point can be considered.
_{Td, CA} = _{Td, AN} = Tin _{, i}

In step S2, a cell outlet temperature _{T out,} the cathode inlet dew point of each single cell _{2 i} from _{i T d, CA} and the anode inlet dew point _{T d,} may be calculated _AN. According to another embodiment, a dew point meter may be used. For example, when no humidifier is used in the fuel cell system 1 or when the cell inlet temperature Tin _{, i} is not used, the stack inlet (the anode side supply port 11a and the cathode side supply port 12a) of the fuel cell 1 is used. It is also possible to install a dew point meter in each of these and set the detected values as the cathode inlet dew point _{Td, CA} and the anode inlet dew point _{Td, AN} . By doing so, estimation with higher accuracy becomes possible.

Further, in an air-based non-humidified system in which the humidifier 30 is not mounted in the air piping system 300, the cathode inlet dew point T _d , _CA can be calculated as 0 ° C. Alternatively, the outside temperature and an external humidity sensor, it is possible to the cathode inlet dew point T _d, the _CA calculated by the outside air temperature and an external humidity of the function. That is, this estimation method can be applied to a non-humidified system.

In step S3 shown in FIG. 8, the water movement speed V _H2O between the electrodes 24A and 24B _{, CA → AN} is obtained. The water moving speed V _{H2O, CA → AN} is calculated by the following equation.
_{VH2O, CA-> AN} = _DH2O * ( _{PH2O, CA- PH2O, AN} )

_{Here, P H2 O, CA} is a water vapor partial pressure of the electrode 24B side of the unit cell _{2 i,} the cathode inlet dew point _{T d,} is calculated by the _CA. PH _{2 O, AN} is the water vapor partial pressure on the electrode 24A side of the single cell 2 _i and is calculated from the anode inlet dew point T _{d, AN} . _{DH 2 O} is a water diffusion coefficient in the electrolyte membrane 23. D _{H2 O,} which can also be used fixed values, because it is intended to change by humidity, they are desirable to consider this.

For example, a characteristic map representing the relationship between the relative humidity of the electrolyte membrane 23 and _{DH 2 O} as shown in FIG. 10 is created in advance, and the value of _{DH 2 O} corresponding to the relative humidity of the electrolyte membrane 23 is created using this characteristic map. Can be used. Specifically, the relative humidity α of the electrolyte membrane 23 estimated at the time of shutdown of the fuel cell stem 1 in the previous operation, the relative humidity α of the electrolyte membrane 23 estimated when the fuel cell stem 1 is left (stopped), or the fuel cell. Using the relative humidity α of the electrolyte membrane 23 estimated immediately before the current estimation in the stem 1, the value (β) of _{DH 2 O} used for the current estimation can be determined from the map.

In step S4 shown in FIG. 8, the water movement speed _{VH2O, CA → AN} , dew point _{Td, CA} , dew point _{Td, AN} , temperature _{TOUT, i} , air back pressure _{Pair, i} , hydrogen pressure _{PH2, i} From the air flow rate Q _{air, i} , the hydrogen flow rate Q _{H2, i} and the current value I, a current density i _x (where x is an arbitrary natural number) is calculated using a map. Current density _{i x} is the current density at any area of the cell surface, for example, each area when the x = 4 and _{s 1 ~s 4, I = i} 1 × s 1 + i 2 × s 2 + I ₃ × s ₃ + i ₄ × s ₄ An example of a result of calculating the distribution of the current density i _x shown in FIG. 11.

In step S4, a current distribution and a relative humidity distribution in the cell plane are calculated. Functions I and RH indicating these are expressed as follows. In addition, each parameter ( _{Td, CA} , _{Td, AN} , _{TOUT, i} , _{Pair, i} , _{PH2, i} , _{Qair, i} , _{QH2, i} , _{VH2O, CA of} the functions I and RH _→ Sensitivity to _AN , i _x ) is mapped in advance. Further, the overvoltage distribution in the cell plane may be calculated from these parameters.
_{I = f (T d, CA} , T d, AN, T OUT, i, P air, i, P H2, i, Q air, i, Q H2, i, V H2O, CA → AN, i x)
_{RH = f (T d, CA} , T d, AN, T OUT, i, P air, i, P H2, i, Q air, i, Q H2, i, V H2O, CA → AN, i x)

  FIG. 12 is a diagram illustrating an example of the relative humidity distribution in the cell plane (relative humidity distribution of the reaction gas flow path and the electrolyte membrane) calculated in step S4. In FIG. 12, the flow form of the counter flow is taken as an example in this embodiment so that the flow of hydrogen gas and air is shown in relation to the position in the cell plane. As shown in FIG. 12, the AN channel (hydrogen channel 25A) is oversaturated with a relative humidity exceeding 100% from the hydrogen inlet 27a to the hydrogen outlet 27b, while the CA channel (air channel 25B). Then, the relative humidity is less than 100% on the air outlet 28b side. Moreover, in the electrolyte membrane 23, the center part (center part of the single cell 2) is in a supersaturated state.

In step S5 shown in FIG. 8, for each of the anode side and the cathode side, the degree of supersaturation σ ₁ (relative humidity exceeds 100%) and the degree of unsaturation σ ₂ (relative) from the result of the relative humidity distribution shown in FIG. The liquid water generation rate V _{vap → liq} and the liquid water evaporation rate V _{liq → vap} are calculated by the following equations. This is because V _{vap → liq} and V _{liq → vap} in the hydrogen flow path 25A and the air flow path 25B are calculated in consideration of the change of the water phase (gas phase, liquid phase) in the reaction gas flow path. It was decided.
V _{vap → liq} = k ₁ × σ ₁
V _{liq → vap} = k ₂ × σ ₂
Here, the coefficients k _{1 and} k ₂ are factors due to temperature and water repellency, and are due to the physical properties of the reaction gas flow path. The coefficients k _{1 and} k ₂ are mapped in advance from the experiment.

In step S6 shown in FIG. 8, the water movement speed V_liq in the reaction gas flow path is calculated by the following equation for each of the anode side and the cathode side. Since the liquid water is blown off by the flow of the reaction gas in the reaction gas channel and is discharged from the cell surface, the water movement speed V_liq in each of the hydrogen channel 25A and the air channel 25B is set in consideration of this. It is decided to calculate.
V_liq = k ₃ × V_gas

Here, the water movement speed V_liq is the movement speed of the liquid water blown off by the reaction gas. V_gas is a water vapor flow rate in the reaction gas flow path, and is calculated from a map relating to a state quantity such as a reaction gas supply flow rate or a water vapor partial pressure. Coefficient k ₃ is the factor with temperature or water repellency is due to the physical properties of the reaction gas channel. Coefficient k ₃ is mapped in advance from experiments.

FIG. 13 is a diagram illustrating an example of the residual water amount distribution in the cell plane calculated through steps S4 to S6. This residual water amount distribution is based on the relative humidity distribution (FIG. 12) of the reaction gas flow path calculated in step S4, and the change in liquid water in the reaction gas flow path (that is, V _{vap → liq} calculated in steps S5 and S6). , V _{liq → vap} and V_liq). As understood from FIG. 13, in the hydrogen channel 25A, the hydrogen outlet 27b side has a larger amount of residual water than the hydrogen inlet 27a side, and in the air channel 25B, the residual water amount gradually decreases toward the air outlet 28b side. ing. Although not shown as a drawing, the moisture content distribution in the cell plane can be obtained from the relative humidity distribution (FIG. 12) of the electrolyte membrane 23 calculated in step S4, and approximated to this relative humidity distribution. It will be a thing.

By the procedure described above, the amount of residual water and the change in water content (water balance) of the single cell 2 _i in a certain calculation time can be calculated, the residual water amount distribution in the hydrogen passage 25A, the residual water amount distribution in the air passage 25B, and the electrolyte. The moisture content distribution of the membrane 23 can be obtained. It is possible to calculate the water balance with the roughness of sensitive meshes (for example, five meshes shown in FIG. 13) in the cell plane, and accurately estimate how much residual water content and water content exist in which part. be able to.

2. _T IN for temperature variation, flow distribution variation in consideration each single cell _{2 i} in the cell lamination direction during _{estimation, i, T OUT, i,} P air, i, P H2, i, Q air, i and _{Q H2, i} Is obtained as follows.

(1) Calculation of Cell Inlet Temperature T _{IN, i As} shown in FIG. 14, first, the stack inlet temperature T _in , the refrigerant flow rate Q _LLC , the outside air temperature T _{outside air} , and the vehicle speed V _{vehicle speed} are read (step S11). Here, T _in is a value detected by the temperature sensor T1. Q _LLC is the flow rate of the refrigerant supplied to the fuel cell 1 and can be estimated from the rotational speed of the cooling pump 50 and other detected values. Alternatively, a flow rate sensor may be provided in the refrigerant flow path 51, and a detection value obtained by the flow rate sensor may be used. T _{outside air} is a detected value by the outside air temperature sensor 62, and V _{vehicle speed} is a detected value by the vehicle speed sensor 63.

In general, in the cell stack 3, the heat radiation increases as the distance from the reactant gas supply port 14 increases, that is, as the cell channel “i” increases. Moreover, the influence of heat dissipation changes according to the refrigerant flow rate, the outside air temperature, and the vehicle speed. For example, as shown in FIG. 15A, the larger the refrigerant flow rate Q _LLC (Q ₁ > Q ₂ ), the less the stack inlet temperature _TIN is affected by heat dissipation. That is, the cell inlet temperature T _{IN, i} need not be lower than the stack inlet temperature T _IN . Further, as shown in FIG. 15B, the higher the T _{outside air} is (T _{outside air 1} > T _{outside air 1} ), the less the stack inlet temperature _TIN is affected by heat radiation.

Therefore, the cell inlet temperature TIN _{, i} is calculated as the following function in consideration of such a decrease in the refrigerant temperature due to heat dissipation (step S12).
T _{IN, i} = f (Q _LLC , T _IN , T _{outside air} , V _{vehicle speed} )
As a result, the cell inlet temperature T _{IN, i} corresponding to the cell channel i can be obtained from the above values of Q _LLC , T _IN , T _{outside air,} and vehicle speed.

(2) Calculation of air flow rate Q _{air, i} and air back pressure P _{air, i As} shown in FIG. 16, first, air flow rate Q _air , air back pressure P _air , stack inlet temperature T _IN , stack outlet temperature T _OUT and current value I are read (step S21). Here, the air flow rate Q _air , the air back pressure P _air and the stack outlet temperature T _OUT are values detected by the flow rate sensor F1, the pressure sensor P1 and the temperature sensor T2, respectively. In step S21, the gas density of the air flowing into the manifold 15a is calculated as a function of the stack inlet temperature T _IN and the air flow rate Q _air .

In the next step S22, based on the residual water content in the single cell 2 _i, determining the P-Q characteristic of the single cell 2 _i (characteristic representing the relationship between the air back pressure and air flow rate). For example, as shown in FIG. 17, a map showing PQ characteristics (pressure-flow rate characteristics) corresponding to a plurality of remaining water amounts (x> y) is acquired in advance, and immediately before the calculation shown in the flow shown in FIG. PQ characteristics corresponding to the remaining water amount (total amount of cathode-side residual water amount of the single cell 2 _i ) are determined.

Next, as a function of the air flow rate Q _air , the air back pressure P _air , the stack outlet temperature T _OUT , the gas density calculated above, and the PQ characteristics of each single cell 2 _i , the cell inlet pressure distribution, the cell inflow flow rate distribution The cell outlet pressure distribution is calculated from the map (step S23). An example of these is as shown in FIGS. Here, the cell inflow flow rate shown in FIG. 18B and the cell outlet pressure shown in FIG. 18C correspond to the air flow rate Q _{air, i} and the air back pressure P _{air, i} corresponding to the cell channel i. Yes (step S24).

Although not described in detail, the hydrogen flow rate Q _{H2, i} and the hydrogen pressure P _{H2, i} of the single cell 2 _i are also calculated by the same method as the calculation of the air flow rate Q _{air, i} and the air back pressure P _{air, i.} can do. In this case, the cell inlet pressure shown in FIG. 18A corresponds to the hydrogen pressure _{PH2, i} , and the cell inflow rate shown in FIG. 18B corresponds to the hydrogen flow rate _{QH2, i} .

(3) Calculation of cell outlet temperature T _{OUT, i As} shown in FIG. 19, first, the stack outlet temperature T _OUT is read as the detection value of the temperature sensor T2 (step S31). Further, as in the case of the stack inlet temperature T _IN described above, the refrigerant flow rate Q _LLC , the outside air temperature T _{outside air} , and the vehicle speed V _{vehicle speed} are read. Further, reads the cell voltage _{V i} and the current value I, to estimate the heat generation amount _{Q cell, i} of each single cell _{2 i} from the I-V characteristic of each unit cell _{2 i.}

Here, the voltage value of each single cell 2 _i detected by a cell monitor (not shown) can be used as the cell voltage V _i . However, instead of using a sensor such as a cell monitor, the cell voltage V _i can be set by providing each single cell 2 _i with an IV map (dependent on power generation amount, air flow rate, hydrogen flow rate, air back pressure, hydrogen pressure). It can also be estimated. The calorific value Q _{cell, i} is due to heat generation due to TΔS and heat loss due to overvoltage.

Similar to the stack inlet temperature T _IN described above, the stack outlet temperature T _OUT is affected by heat radiation depending on the position of the single cell 2 _i in the cell stack 3. For example, as shown in FIG. 20, the stack outlet temperature T _OUT is less affected by heat dissipation as the refrigerant flow rate Q _LLC increases (Q _LLC1 <Q _LLC2 ).

Therefore, the cell outlet temperature T _{OUT, i} is calculated as the following function in consideration of the heat generation amount Q _{cell, i} , the refrigerant flow rate Q _{LLC, i} and the heat radiation (step S32).
T _{OUT, i} = f (Q _{cell, i} , Q _{LLC, i} , T _OUT , T _{outside air} , V _{vehicle speed} )
Thereby, the cell outlet temperature T _{OUT, i} corresponding to the cell channel i can be obtained from each detected value or estimated value indicated by these parameters.

Note that Q _{LLC, i} is the refrigerant flow rate supplied to each single cell 2, and the distribution amount of the refrigerant flow rate Q _LLC when considering the fuel cell stack 1 as one point is taken into consideration. Specifically, a map representing the relationship between the refrigerant flow rate Q _LLC and the cell channel i is created in advance for each of the several refrigerant flow rates Q _LLC , and this map is used to select the Q corresponding to the cell channel i. _{LLC, i} can be calculated.

According to the procedure of described above (1) to (3), in the flow (steps S1, S2 and S4) shown in FIG. 8, the temperature distribution (heat radiation amount in the cell lamination direction about the state of each single cell 2 _i A value that takes into account the distribution of pressure loss and the distribution of pressure loss (oxidation gas, fuel gas, refrigerant distribution, etc.). Thereby, compared with the case where the fuel cell 1 is grasped by one point as a stack, the residual water amount distribution and the water content distribution can be accurately estimated for all the single cells 2 (that is, in the cell stacking direction).

D. Example of estimation while system is left and control result using estimation result First, with reference to FIG. 21 to FIG. 24, general fluctuations in the amount of remaining water in the pattern of operation stop, leaving and restart of the fuel cell system 100 will be described. Next, with reference to FIG. 25 and FIG. 26, estimation of the anode side residual water amount during the leaving period of the fuel cell system 100 will be described. And finally, with reference to FIG. 27A and B, the example of control using the estimation result in the leaving period is demonstrated.

FIG. 21 is a diagram illustrating an example of a change in the remaining water amount of the single cell 2 over time when the control example according to the present embodiment is not executed.
As shown in FIG. 21, the remaining amount of water on the cathode side is larger than the remaining amount of water on the anode side in the section during operation of the fuel cell system 100 (until time t ₁ ). This is because, as described above, water is generated along with the power generation of the single cell 2 on the cathode side. If you stop the operation of the fuel cell system 100 in a time t _1, the subsequent standing period (time t ₁ ~t _2), water to the anode side from the cathode side by the water vapor partial pressure difference (temperature difference) moves, residual water The size of can be reversed. Further, the amount of remaining water on the anode side may be larger at the time of system restart (time t ₂ ) than at the time of system stop (time t ₁ ).

22A and 22B show an example of the residual water amount distribution and the pressure loss distribution in the hydrogen flow path 25A at the time of system restart (time t ₂ ) when the cell stack 3 shown in FIG. 9 is used as a model.
As shown in FIG. 22A, the end cell 2a has a larger amount of residual water, and the single cell 2 in the center in the cell stacking direction (hereinafter referred to as the main cell 2b; see FIG. 9) has a small amount of residual water. I understand. In particular, the end cell 2a on the far side in the cell stacking direction (downstream in the reaction gas supply direction in the cell stack 3) has a larger amount of residual water than the end cell 2a on the near side in the cell stacking direction. As the amount of remaining water increases, the flow of hydrogen gas is hindered, and the pressure loss in the hydrogen flow path 25A increases (see FIG. 22B).

FIG. 23 shows an example of a change in cell temperature while the system is left when the cell stack 3 shown in FIG. 9 is used as a model.
As shown by curves L _{1 to} L ₃ in FIG. 23, the cell temperature greatly decreases as the standing period (= t ₂ −t ₁ ) of the fuel cell system 100 becomes longer. The cell temperature varies depending on the position of the single cell 2 in the cell stack 3, and the cell temperature of the end cell 2a is the lowest. This is because the heat radiation amount is larger at the end of the cell stack 3. The cell temperature is, for example, the above-described cell outlet temperature T _{OUT, i} .

FIG. 24 shows an example of the relationship between the temperature of the single cell 2 and the saturated water vapor pressure while the system is left standing.
As shown in FIG. 24, in any single cell 2, the temperature T _AN of the anode electrode 24A on the outer side in the cell stacking direction tends to be lower than the temperature T _CA of the cathode electrode 24B on the inner side in the cell stacking direction. This temperature difference ΔT tends to be larger in the end cell 2a than in the main cell 2b. This is due to the difference in the amount of heat dissipation described above. Due to the temperature difference ΔT, a difference in water vapor partial pressure occurs between the electrodes 24A and 24B, and water moves through the electrolyte membrane 23 from the cathode electrode 24B side to the anode electrode 24A side. The greater the temperature difference ΔT, the greater the difference in water vapor partial pressure, so the amount of water movement increases.

  As a result of water movement while the system is left, there is a risk that the supply amount of hydrogen gas to the hydrogen flow path 25A is insufficient when the system is restarted. In particular, when there is a rapid output increase request such as WOT (Wide Open Throttle) when the fuel cell system 100 is mounted on a vehicle, there is a high possibility that the supply amount of hydrogen gas will be insufficient. If the supply amount becomes insufficient, the MEA 20 may be deteriorated. In order to suppress the shortage of hydrogen gas supply at the time of system restart, it is desirable to accurately grasp the water movement while the system is left and maintain the remaining water amount in the hydrogen flow path 25A in an appropriate state.

Therefore, in the present embodiment, the residual water amount distribution in the hydrogen flow path 25A is estimated even when the system is left, and for this purpose, the temperature of the single cell 2 is acquired while the system is left. In this case, the cell outlet temperature T _{OUT, i} described above is acquired as the temperature of the single cell 2. That is, in the present embodiment, by using the method described in “(3) Calculation of cell outlet temperature T _{OUT, i} ”, one stack outlet temperature T _OUT is detected by the temperature sensor T2, and the estimation unit 67 Calculates the cell outlet temperature T _{OUT, i} of each single cell 2 from this detected value.

Here, in the calculation of the cell outlet temperature T _{OUT, i} during the system operation described above, the following function was used.
T _{OUT, i} = f (Q _{cell, i} , Q _{LLC, i} , T _OUT , T _{outside air} , V _{vehicle speed} )
However, in calculating the cell outlet temperature T _{OUT, i} while the system is left, the parameters of the calorific value Q _{cell, i} , the refrigerant flow rate Q _{LLC, i} , and the V _{vehicle speed} may not be used. This is because these values are zero when the system is left unattended.
Therefore, as the cell outlet temperature T _{OUT, i} while the system is left standing, a value calculated from the following function considering the variation of the heat radiation amount in the cell stack 3 is used.
T _{OUT, i} = f (T _OUT , T _{outside air} )

Here, the number of times the cell outlet temperature T _{OUT, i} is calculated while the system is left is at least once, but preferably a plurality of times. The reason for this will be described with reference to FIG.

FIG. 25 shows the time change of the cell outlet temperature T _{OUT, i} from the system stop to the system restart (that is, the system is left). Depending on the conditions under which the fuel cell system 100 is placed _, the temperature drop profile of the cell outlet temperature T _{OUT, i} while the system is left is different. For this reason, the temperature drop profile 1 indicated by the dotted line in FIG. 25 and the temperature drop profile 2 also indicated by the solid line differ in the manner of the temperature distribution in the cell stack 3, and the hydrogen flow path 25A due to the temperature difference ΔT between the electrodes. The amount of water movement inwards will be different.

In this regard, if an attempt is made to estimate only the temperature information of the cell outlet temperature T _{OUT, i} when the system is stopped and the system is restarted _, what kind of temperature drop profile does the cell outlet temperature T _{OUT, i} have while the system is left alone? I can't figure out what to follow. As a result, the estimation accuracy of the residual water amount distribution in the hydrogen flow path 25 is deteriorated. In addition, when the cell outlet temperature T _{OUT, i} is calculated only once while the system is left, it is possible to grasp a certain temperature drop profile, but in order to grasp with higher accuracy, the cell outlet temperature T _{OUT, i} Is preferably calculated a plurality of times.

Therefore, as shown in FIG. 26, in the present embodiment, the cell outlet temperature T _{OUT, i} of each single cell 2 is calculated a plurality of times while the system is left (see: plots A _{1 to} A ₃ ). From the calculation result, a temperature profile indicating a temperature change of the cell outlet temperature T _{OUT, i} while the system is left is acquired with high accuracy. Then, the transition of the temperature distribution while the system is left standing is estimated from the acquired temperature profile, and the amount of residual water on the anode side when the system is left standing by using the above-mentioned “1. Method for estimating the water distribution in the cell plane”. The distribution of water content including the distribution is estimated.

  Therefore, according to the estimation of the anode side residual water amount while the system is left according to the present embodiment described above, it is possible to accurately grasp the water movement while the system is left. The amount of remaining water can be estimated accurately. Thereby, countermeasure control (for example, a scavenging process described later) according to the amount of remaining water in the hydrogen flow path 25A can be accurately performed.

Note that the estimated value (cell outlet temperature T _{OUT, i} is acquired and the remaining water amount in the hydrogen flow path 25A based on the estimated time is obtained only during a temporary period from system stop to system restart using a timer while the system is left. Estimation).

  Next, with reference to FIGS. 27A and 27B, a control example using the above estimation result of the remaining water amount in the hydrogen flow path 25A while the system is left will be described. In this control example, the scavenging process is performed on the hydrogen flow path 25A when the amount of remaining water in the hydrogen flow path 25A estimated during monitoring of the temperature profile reaches the predetermined threshold. It is. The control related to the scavenging process is executed by the operation control unit 68 that receives the signal from the estimation unit 67.

The vertical axis of FIG. 27A shows the residual water amount of the single cell 2, and the figure shows the anode residual water amount (estimated value of the total residual water amount in the hydrogen flow path 25 </ b> A) and the cathode residual water amount (air flow) in the single cell 2. The transition of the total amount of remaining water in the path 25B) is shown. Residual water content W ₁ is a threshold for determining whether to execute the scavenging process, shows the water clogging occurs residual water content of the hydrogen flow path 25A.

As shown in FIG. 27A, since the water movement is performed as described above while the system is left, the amount of remaining cathode water decreases with time, while the amount of remaining anode water increases accordingly. In this control example, as shown in FIG. 27B, when the amount of remaining anode water reaches the amount of remaining water W ₁ (time t ₁ ), the hydrogen pump 46 is rotated for a certain period (= time t ₂ -t ₁ ). The scavenging process is performed. In this case, the anode residual water content of comparing whether reaches residual water content W ₁ may be any anode residual water content of at least one single cell 2 or end cell 2a. And by such a scavenging process, the gas pumped from the hydrogen pump 46 will carry away the liquid water in the hydrogen flow path 25A, and the amount of anode residual water will fall. Thereafter, as shown at time t ₃ , if the remaining anode water amount reaches the remaining water amount W ₁ again, the hydrogen pump 46 is rotated again to perform the scavenging process.

  As described above, according to the control example of the present embodiment, since the highly accurate estimation result of the water distribution is used, the scavenging process can be appropriately performed while the system is left. Thereby, since it is not necessary to take measures against the excessive residual water amount when the system is restarted, it is possible to contribute to improvement of the user's drivability. Further, it is not necessary to rotate the hydrogen pump 46 for the scavenging process when the system is restarted, and noise can be reduced, so that NV can be improved. Furthermore, when the system is restarted, insufficient supply of hydrogen gas to the hydrogen flow path 25A can be suppressed, leading to improved fuel efficiency. On the other hand, by performing the scavenging process while the system is left, it is possible to shorten the scavenging process time when the scavenging process is performed when the system is stopped and when the system is restarted.

  In another embodiment, the scavenging process can be performed by supplying an inert gas (for example, nitrogen) to the hydrogen flow path 25A.

1 is a perspective view of a fuel cell according to an embodiment. It is a side view which shows a part of inside of the fuel cell which concerns on embodiment. It is sectional drawing of the single cell which concerns on embodiment. It is a top view of the separator which concerns on embodiment. It is a schematic plan view which shows the flow-path shape of the separator which concerns on the 1st modification of embodiment. It is a schematic plan view which shows the flow-path shape of the separator which concerns on the 2nd modification of embodiment. It is a schematic plan view which shows the flow-path shape of the separator which concerns on the 3rd modification of embodiment. 1 is a configuration diagram of a fuel cell system according to an embodiment. It is a functional block diagram of a control device concerning an embodiment. It is a flowchart which shows the estimation method of the water distribution in the cell surface which concerns on embodiment. It is a figure which shows the relationship between supply and discharge | emission of a reactive gas, and a cell channel about the cell laminated body which concerns on embodiment. It is a characteristic map showing the relationship between the relative humidity of the electrolyte membrane which concerns on embodiment, and _DH2O . It is a figure which shows the current density with respect to the position in a cell surface which concerns on embodiment. It is a figure which shows the relative humidity distribution of the reactive gas flow path and electrolyte membrane in the cell surface which concerns on embodiment. It is a figure which shows the residual water amount distribution in the cell surface which concerns on embodiment. It is a flowchart which shows the calculation method of the cell inlet_port | entrance temperature which concerns on embodiment. It is a figure which shows the relationship between the position of a single cell, and a refrigerant | coolant flow rate about the influence of the thermal radiation which the stack inlet_temperature which concerns on embodiment receives. It is a figure which shows the relationship between the position of a single cell, and external temperature about the influence of the heat radiation which the stack | stuck inlet temperature which concerns on embodiment receives. It is a flowchart which shows the calculation method of the air flow rate and air back pressure for every single cell which concerns on embodiment. It is a map which shows the PQ characteristic of the single cell according to the residual water amount which concerns on embodiment. It is a figure which shows the cell inlet pressure distribution which concerns on embodiment. It is a figure which shows the cell inflow flow rate distribution which concerns on embodiment. It is a figure which shows the cell exit pressure distribution which concerns on embodiment. It is a flowchart which shows the calculation method of the cell exit temperature which concerns on embodiment. It is a figure which shows the relationship between the position of a single cell, and a refrigerant | coolant flow rate about the influence of the heat radiation which the stack exit temperature which concerns on embodiment receives. It is a figure which shows an example of the change of the residual water amount of the single cell with the passage of time when not performing the example of control concerning an embodiment. It is a figure which shows an example of residual water amount distribution of the hydrogen flow path at the time of system restart when the cell laminated body shown in FIG. 9 is used as a model. It is a figure which shows an example of the pressure loss distribution of the hydrogen flow path at the time of system restart when the cell laminated body shown in FIG. 9 is used as a model. It is a figure which shows an example of the change of the cell temperature in the case of leaving the system when the cell laminated body shown in FIG. 9 is used as a model. It is a figure which shows an example of the relationship between the temperature of the single cell in which the system is left to stand, and a saturated water vapor pressure when the cell stack shown in FIG. 9 is used as a model. It is a figure used for demonstrating estimation of the anode side residual water amount in the system leaving stand according to embodiment, and is a figure which shows the aspect of the temperature change of the single cell in the system leaving stand. It is a figure which shows the temperature profile of the single cell acquired while the system which concerns on embodiment is left. It is a timing chart which shows the scavenging process in the system leaving stand concerning the example of control of an embodiment, and is a figure showing the amount of remaining water of each cell on the vertical axis. It is a timing chart which shows the scavenging process in the system leaving stand concerning the example of control of an embodiment, and is a figure showing the number of rotations of a hydrogen pump on the vertical axis.

Explanation of symbols

  1: fuel cell, 2: single cell, 2a: main cell, 2b: end cell, 23: electrolyte membrane, 24A: anode electrode, 24B: cathode electrode, 25A: hydrogen channel (fuel gas channel), 25B: Air flow path (oxidation gas flow path), 67: estimation section, 68: operation control section, 100: fuel cell system, 300: air piping system, 400: hydrogen piping system, 500: refrigerant piping system, 600: control device

Claims (4)

A plurality of single cells each having an anode electrode, a cathode electrode, an electrolyte membrane between them, a fuel gas channel for supplying fuel gas to the anode electrode, and an oxidizing gas channel for supplying oxidizing gas to the cathode electrode are stacked. A fuel cell system including a fuel cell,
The water movement performed between the anode electrode and the cathode electrode through the electrolyte membrane, and the temperature of each single cell is calculated a plurality of times during the standing period from the stop to restart of the fuel cell system In consideration of the temperature drop profile of each single cell obtained as described above, an estimation unit that estimates the residual water amount distribution of the fuel gas flow channel in the cell plane of each single cell during the standing period ,
An operation control unit that performs a scavenging process on the fuel gas flow path when a total amount of residual water in the fuel gas flow path ascertained from the estimated residual water flow distribution reaches a predetermined threshold; A fuel cell system provided . A temperature sensor for detecting a temperature related to the fuel cell;
The fuel cell system according to claim 1, wherein the estimation unit calculates the temperature of each single cell using one temperature detected by the temperature sensor. 3. The fuel cell system according to claim 1, wherein the estimation unit estimates a residual water amount distribution of the oxidizing gas flow channel in a cell plane of each single cell in consideration of the water movement. The fuel cell system according to any one of claims 1 to 3 , wherein the estimation unit estimates a water content distribution of the electrolyte membrane in a cell plane of each single cell in consideration of the water movement.

Images